Parking space recommendation method and system based on superior and inferior solution distance method

Abstract

The invention discloses a good and inferior solution distance method-based parking space recommendation method and system, and the method comprises the steps: constructing an evaluation index system of a driving trip scene, determining the attribute preference weight of a parking space through a G1 algorithm, carrying out the sorting, obtaining index parameters, collecting and sorting all kinds of attribute information of a plurality of parking spaces, and forming a parking space multi-attribute evaluation matrix, establishing an upper bound matrix and a lower bound matrix based on the parking space multi-attribute evaluation matrix, respectively determining a positive ideal solution and a negative ideal solution corresponding to each index parameter, comparing each index parameter with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each index parameter, and using an accurate value and an interval value to jointly describe parking space attributes. A G1 method is used to determine each attribute weight according to the preference of a user, an upper bound matrix and a lower bound matrix are established for a complex data scene so as to obtain a positive ideal scheme and a negative ideal scheme, a grey correlation degree is introduced to carry out distance measurement of TOPSIS, a comprehensive evaluation score of each parking space is calculated, a sequence is determined, and appropriate parking spaces are recommended for different users.

Description

Technical Field

[0001] The present invention belongs to the technical field of parking space recommendation, and in particular relates to a parking space recommendation method and system based on a superior-inferior solution distance method. Background Art

[0002] With the improvement of social and economic levels, the number of motor vehicles and parking demand in cities are increasing, but traffic planning and parking space construction are relatively slow, which greatly restricts people's transportation. Intelligent reservation and management of parking spaces can grasp the use of parking spaces in real time, recommend and reserve parking spaces for users, thereby greatly reducing the turnover time of vehicles looking for parking spaces, improving traffic efficiency, and reducing environmental pollution and energy waste. In this way, users can arrange their travel reasonably, and traffic management departments can also achieve macro-control of the overall traffic situation.

[0003] In order to recommend parking spaces to users, it is necessary to conduct horizontal and vertical comparative analysis of various indicators such as parking fees, parking lot safety, and parking convenience, and select the most satisfactory solution from multiple alternatives. This is a typical multi-attribute decision-making problem. Multi-attribute decision-making problems are an important part of modern decision-making theory. Many problems in real life can be attributed to multi-attribute decision-making problems. To solve this problem, researchers have proposed multi-attribute decision-making methods such as entropy weight method, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), Analytic Hierarchy Process (AHP), and Data Envelopment Method.

[0004] However, the above multi-attribute decision-making methods are mostly used for large-scale project evaluation, and the calculation process is relatively complicated, while there are few methods for evaluating parking spaces. Due to the uncertainty of the real world, some attributes of parking spaces cannot give accurate values. The traditional TOPSIS method uses Euclidean distance for measurement, which may ignore the linear relationship between indicators and cannot process interval value data. Based on this, the present invention proposes a parking space recommendation method with uncertain attributes to solve the above-mentioned technical problems. Summary of the invention

[0005] In view of this, the present invention provides a parking space recommendation method and system based on the superior-inferior solution distance method, which uses exact values ​​and interval values ​​to jointly describe parking space attributes, uses the G1 method to determine the weights of each attribute according to user preferences, and for complex data scenarios, establishes upper and lower bound matrices to obtain positive and negative ideal solutions, introduces grey correlation to perform TOPSIS distance measurement, calculates the comprehensive evaluation score of each parking space and determines the ranking, and is specifically implemented using the following technical solutions.

[0006] In a first aspect, the present invention provides a parking space recommendation method based on a superior-inferior solution distance method, comprising the following steps:

[0007] Constructing an evaluation index system for driving travel scenarios, wherein the evaluation index system includes five parking space attributes: time to arrive at the parking space, hourly parking price, parking convenience, walking distance from the parking space to the final destination, and parking lot safety;

[0008] Determine the attribute preference weights of parking spaces using the G1 algorithm according to the evaluation index system, and sort the attribute preference weights to obtain the index parameters of the evaluation index system;

[0009] Collect and organize various attribute information of multiple parking spaces, and form a parking space multi-attribute evaluation matrix based on the various attribute information;

[0010] An upper bound matrix and a lower bound matrix are established based on the parking space multi-attribute evaluation matrix, and a positive ideal solution and a negative ideal solution corresponding to each of the indicator parameters are determined respectively, wherein the positive ideal solution and the negative ideal solution are determined by using a TOPSIS algorithm;

[0011] Each index parameter is compared with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each index parameter, so as to recommend a suitable parking space for the user.

[0012] As a preferred embodiment of the above technical solution, the attribute preference weight of the parking space is determined by using the G1 algorithm according to the evaluation index system, including:

[0013] The differences in the importance that different users attach to parking space attributes are used as attribute preference weights, and the attribute preference weights are determined using the G1 algorithm;

[0014] The evaluation index system is sorted, and the index parameters after sorting are recorded as I 1 ≥I 2 ≥,...≥I j-1 ≥I j , where I i ≥I j Indicates that the importance of indicator i is greater than or equal to the importance of indicator j;

[0015] Defining Indicators I i-1 with I i The importance ratio is r i , then:

[0016]

[0017] Among them, ω i represents the weight of the i-th indicator;

[0018] Summing k from 2 to n, we get:

[0019]

[0020] according to Calculate the relative weight of the last indicator as:

[0021]

[0022] Finally, we get ω through formula (4): i-1 The weights of the indicators are:

[0023] ω i-1 =r i ω i-1 (4)

[0024] Among them, i=2,3,...n-1,n.

[0025] As a preferred embodiment of the above technical solution, various attribute information of multiple parking spaces is collected and sorted, and a parking space multi-attribute evaluation matrix is ​​formed according to the various attribute information, including:

[0026] Collect and organize m types of attribute information of n parking spaces to form a parking space multi-attribute evaluation matrix A = (a ij ) m×n ;

[0027] The data in the parking space multi-attribute evaluation matrix is ​​normalized to obtain a standardized matrix Z=(z ij ) m×n ;

[0028] When data a ij When is the exact value,

[0029] When data a ij When is an interval value, After data processing, we get

[0030] Among them, z ij represents the normalized data. represents the lower bound data after normalization, represents the upper bound data after normalization, Represents parking space multi-attribute information a ij The lower limit of Represents parking space multi-attribute information a ij The upper limit of

[0031]

[0032] As a preferred embodiment of the above technical solution, the data in the parking space multi-attribute evaluation matrix is ​​normalized, including:

[0033] Select the time U to arrive at the parking space 1 Hourly parking fee U 2 , Parking convenience 3 , walking distance from the parking space to the final destination U 4 and parking lot safety 5 ;

[0034] Introduce interval value to describe the time U to arrive at the parking space 1 and parking convenience 3 There is uncertainty, so an exact value is used to describe the hourly parking price U 2 , walking distance from the parking space to the final destination U 4 and parking lot safety 5 , parking convenience 3 and parking lot safety 5 Experts will rate the results on a scale of 1-9.

[0035] As a preferred embodiment of the above technical solution, a parking convenience of 1 indicates the most inconvenient parking, and a parking convenience of 9 indicates the most convenient parking. The driving travel scenarios include at least one of road driving, cruising to select a parking space, parking, and walking to the destination.

[0036] As a preferred embodiment of the above technical solution, an upper bound matrix and a lower bound matrix are established based on the parking space multi-attribute evaluation matrix, and the positive ideal solution and the negative ideal solution corresponding to each of the indicator parameters are determined respectively, including:

[0037] Establish the upper bound matrix Z U With the lower bound matrix Z L , and determine the positive ideal solution and negative ideal solution Z L ;

[0038] The grey decision analysis algorithm is used to calculate the difference between each parking space and the ideal solution. Negative ideal solution Z L distance;

[0039] Among them, the TOPSIS algorithm needs to first determine the positive ideal solution and the negative ideal solution, and then compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain the score of each indicator parameter.

[0040] As a preferred embodiment of the above technical solution, the TOPSIS algorithm needs to first determine the positive ideal solution and the negative ideal solution, including:

[0041] Upper Bound Matrix and the lower bound matrix According to the standardized matrix Z = (z ij ) m×n Calculation results in formulas (8) and (9):

[0042]

[0043] In the upper bound matrix Z U In, order is a positive ideal solution, where is the maximum value among the i-th index;

[0044] In the lower bound matrix Z L In, order z L ={ z 1 , z 2 ,... z m} T is a negative ideal solution, where z i is the minimum value of the i-th index, and T is the transposition operation.

[0045] As a preferred embodiment of the above technical solution, a grey decision analysis algorithm is used to calculate the distance between each parking space and the positive ideal solution and the negative ideal solution, and to sort the scores of each indicator parameter, including:

[0046] Upper bound matrix Z U The grey correlation matrix between the positive ideal solution is The elements in the matrix are:

[0047]

[0048] Lower bound matrix Z L The grey correlation matrix between the negative ideal solution is The elements in the matrix are:

[0049]

[0050] Wherein, in formula (10) and formula (11), ρ represents the gray resolution coefficient;

[0051] The weighted grey correlation between each parking space and the positive ideal solution and the negative ideal solution calculated according to the grey correlation matrix is: The improved distance is:

[0052]

[0053] Among them, ω in formula (12) jis the attribute preference weight obtained in formula (4).

[0054] As a preferred embodiment of the above technical solution, the relative closeness T between the parking space and the positive ideal solution and the negative ideal solution is calculated. j As a quantitative scoring result, the relative closeness calculation formula is:

[0055]

[0056] When the relative closeness approaches 0, the closer the parking space is to the negative ideal solution, the lower the score; when the relative closeness approaches 1, the closer the parking space is to the positive ideal solution, the higher the score. The ranking of each parking space is obtained and a suitable parking space is recommended to the user.

[0057] In a second aspect, the present invention further provides a parking space recommendation system based on the superiority-inferior solution distance method, which is applied to the above-mentioned parking space recommendation method based on the superiority-inferior solution distance method, comprising:

[0058] An index system construction module is used to construct an evaluation index system for driving travel scenarios, wherein the evaluation index system includes five parking space attributes, namely, the time to arrive at the parking space, the hourly parking price, the convenience of parking, the walking distance from the parking space to the final destination, and the safety of the parking lot;

[0059] An indicator parameter determination module, used to determine the attribute preference weights of parking spaces using the G1 algorithm according to the evaluation indicator system, and to sort the attribute preference weights to obtain indicator parameters of the evaluation indicator system;

[0060] An evaluation matrix forming module, used to collect and organize various attribute information of multiple parking spaces, and form a parking space multi-attribute evaluation matrix based on the various attribute information;

[0061] A positive and negative ideal solution determination module, used to establish an upper bound matrix and a lower bound matrix based on the parking space multi-attribute evaluation matrix, and respectively determine the positive ideal solution and the negative ideal solution corresponding to each of the indicator parameters, wherein the positive ideal solution and the negative ideal solution are determined by using a TOPSIS algorithm;

[0062] The parking space recommendation module is used to compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each indicator parameter, so as to recommend a suitable parking space for the user.

[0063] The invention provides a parking space recommendation method and system based on the superior-inferior solution distance method. The method comprises the following steps: constructing an evaluation index system for a driving travel scenario, adopting a G1 algorithm to determine the attribute preference weights of a parking space according to the evaluation index system, and sorting the attribute preference weights to obtain the index parameters of the evaluation index system. Various attribute information of a plurality of parking spaces is collected and sorted, and a parking space multi-attribute evaluation matrix is ​​formed according to the various attribute information. An upper bound matrix and a lower bound matrix are established based on the parking space multi-attribute evaluation matrix, and positive ideal solutions and negative ideal solutions corresponding to each index parameter are respectively determined. Each index parameter is compared with the positive ideal solution and the negative ideal solution to obtain the score corresponding to each index parameter, so as to recommend a suitable parking space to a user. The parking space attributes are described together with an exact value and an interval value. The G1 method is used to determine the weights of each attribute according to the user's preference. For complex data scenarios, upper and lower bound matrices are established to obtain positive and negative ideal solutions. Grey correlation is introduced to perform TOPSIS distance measurement, and the comprehensive evaluation score of each parking space is calculated and the sorting is determined. The method can recommend a suitable parking space to different types of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0065] Figure 1 It is a flow chart of the parking space recommendation method based on the superior and inferior solution distance method of the present invention;

[0066] Figure 2 It is a structural block diagram of the parking space recommendation system based on the superior-inferior solution distance method of the present invention. DETAILED DESCRIPTION

[0067] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0068] According to the uncertainty of some attributes of parking spaces and user preferences, the present invention proposes a new parking space recommendation method based on improved TOPSIS. The method first constructs the deterministic and uncertain attributes of parking spaces, which are described by exact values ​​and interval values ​​respectively. Then, the G1 method is used to determine the weight of each attribute according to the user's preference. For complex data scenarios containing exact values ​​and interval values, upper and lower bound matrices are established to obtain positive and negative ideal solutions. Finally, grey correlation is introduced to improve the distance metric of TOPSIS so as to comprehensively evaluate the alternative parking spaces and give the optimal ranking. Case studies show that the proposed method can recommend suitable parking spaces according to user preferences.

[0069] See also Figure 1 ,The parking space recommendation method based on the superior and inferior solution distance method includes the following steps:

[0070] S1: Construct an evaluation index system for driving travel scenarios, wherein the evaluation index system includes five parking space attributes, namely, time to arrive at the parking space, hourly parking price, parking convenience, walking distance from the parking space to the final destination, and parking lot safety;

[0071] S2: Determine the attribute preference weights of parking spaces using the G1 algorithm according to the evaluation index system, and sort the attribute preference weights to obtain index parameters of the evaluation index system;

[0072] S3: collecting and collating various attribute information of multiple parking spaces, and forming a parking space multi-attribute evaluation matrix according to the various attribute information;

[0073] S4: establishing an upper bound matrix and a lower bound matrix based on the parking space multi-attribute evaluation matrix, and respectively determining a positive ideal solution and a negative ideal solution corresponding to each of the indicator parameters, wherein the positive ideal solution and the negative ideal solution are determined by using a TOPSIS algorithm;

[0074] S5: Compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each indicator parameter, so as to recommend a suitable parking space for the user.

[0075] In this embodiment, the driving travel scenario consists of the following activities: road driving, cruising to select a parking space, parking, and walking to the destination. Parking lot entrances and exits are prone to congestion. When there are multiple parking lots near the destination, the driver will estimate the time period to reach the parking space and make a choice based on his own situation; in large parking lots, after parking, it is still necessary to walk a distance to reach the final destination. Some drivers will pay attention to the length of the walking distance. The safety level of different parking lots is also different. For example, a parking lot guarded by security guards is safer than an open-air unmanned parking lot; even in the same parking lot, the safety level of different parking spaces is different. For example, parking spaces near the entrances and exits of parking lots or entrances and exits on different floors are prone to scratches; similarly, the convenience level of parking in different parking spaces is also different. After communicating with experts and drivers, it was found that the time U to reach the parking space is different. 1 and parking convenience 3 These two attributes are uncertain because they cannot be estimated. In order to accurately describe them, interval values ​​are introduced to describe them. The hourly parking price U 2 , walking distance from the parking space to the final destination U 4 and parking lot safety 5 The convenience of parking can be described by an exact value, U 3 and parking lot safety 5 Experts will rate the property on a scale of 1-9, with the larger the value, the better the property. For example, a parking convenience of 1 means the least convenient parking, and a parking convenience of 9 means the most convenient parking. Other data can be obtained through actual measurements.

[0076] It should be noted that the data in the parking space multi-attribute evaluation matrix is ​​normalized, including: selecting the time U to arrive at the parking space 1 Hourly parking fee U 2 , Parking convenience 3 , walking distance from the parking space to the final destination U 4 and parking lot safety 5 ; Introduce interval value to describe the time U to arrive at the parking space 1 and parking convenience 3 There is uncertainty, so an exact value is used to describe the hourly parking price U 2 , walking distance from the parking space to the final destination U 4 and parking lot safety 5 , parking convenience 3 and parking lot safety 5Experts will give scores on a scale of 1-9. Intelligent parking reservation and management is an important part of smart cities. Its advanced experience can be promoted in other fields such as smart business districts, smart communities, and smart campuses, driving the informatization construction of related fields; it will also help local industrial upgrading and transformation. The parking space recommendation problem is undoubtedly the core and difficulty of intelligent parking reservation and management. In view of the uncertainty of certain attributes of parking spaces, it is proposed to combine precise values ​​with interval values ​​to jointly describe parking space information, and then establish a new parking space recommendation method based on improved TOPSIS. Starting from the user's perspective, this method uses the G1 method to determine the attribute weights; establish upper and lower bound matrices respectively to obtain positive and negative ideal solutions; introduce grey correlation measurement to optimize TOPSIS, and calculate the difference between each parking space and the positive ideal solution and negative ideal solution Z L Finally, the alternative parking spaces are comprehensively evaluated and ranked. The actual example shows that this method has the advantages of being comprehensive, convenient and accurate, and can recommend suitable parking spaces for different types of users.

[0077] Optionally, the attribute preference weight of the parking space is determined using a G1 algorithm according to the evaluation index system, including:

[0078] The differences in the importance that different users attach to parking space attributes are used as attribute preference weights, and the attribute preference weights are determined using the G1 algorithm;

[0079] The evaluation index system is sorted, and the index parameters after sorting are recorded as I 1 ≥I 2 ≥,...≥I j-1 ≥I j , where I i ≥I j Indicates that the importance of indicator i is greater than or equal to the importance of indicator j;

[0080] Defining Indicators I i-1 with I i The importance ratio is r i , then:

[0081]

[0082] Among them, ω i represents the weight of the i-th indicator;

[0083] Summing k from 2 to n, we get:

[0084]

[0085] according to Calculate the relative weight of the last indicator as:

[0086]

[0087] Finally, we get ω through formula (4): i-1 The weights of the indicators are:

[0088] ω i-1 =r i ω i-1 (4)

[0089] Among them, i=2,3,...n-1,n.

[0090] In this embodiment, in real scenarios, different users have different degrees of attention to parking space attributes. This difference in attention is called attribute preference weight, which must be able to reflect the impact of the indicator system on the parking space recommendation scheme to the greatest extent. To this end, the G1 algorithm is used to determine the attribute preference weight. The G1 algorithm is a subjective weighting method proposed on the basis of AHP. Compared with AHP, the biggest advantage of the G1 algorithm is that it does not need to construct a discriminant matrix and consistency test, and the amount of calculation is small. Among them, r i The values ​​of are shown in Table 1:

[0091] Table 1r i Assignment correspondence table

[0092]

[0093] Optionally, various attribute information of multiple parking spaces is collected and sorted, and a parking space multi-attribute evaluation matrix is ​​formed according to the various attribute information, including:

[0094] Collect and organize m types of attribute information of n parking spaces to form a parking space multi-attribute evaluation matrix A = (a ij ) m×n ;

[0095] The data in the parking space multi-attribute evaluation matrix is ​​normalized to obtain a standardized matrix Z=(z ij ) m×n ;

[0096] When data a ij When is the exact value,

[0097] When data a ij When is an interval value, After data processing, we get

[0098]

[0099] Among them, z ij represents the normalized data. represents the lower bound data after normalization, represents the upper bound data after normalization, Represents parking space multi-attribute information a ij The lower limit of Represents parking space multi-attribute information a ij The upper limit of

[0100]

[0101] In this embodiment, m various attribute information of n parking spaces are collected and sorted to form a parking space multi-attribute evaluation matrix A=(a ij ) m×n , the data in the matrix have different dimensions and large differences in magnitude, and there are both accurate values ​​and interval values. In order to make the data comparable, the following formulas (5) and (6) can be used for normalization to obtain a standardized matrix.

[0102] Optionally, an upper bound matrix and a lower bound matrix are established based on the parking space multi-attribute evaluation matrix, and a positive ideal solution and a negative ideal solution corresponding to each of the indicator parameters are determined respectively, including:

[0103] Establish the upper bound matrix Z U With the lower bound matrix Z L , and determine the positive ideal solution and negative ideal solution Z L ;

[0104] The grey decision analysis algorithm is used to calculate the difference between each parking space and the ideal solution. Negative ideal solution Z L distance;

[0105] Among them, the TOPSIS algorithm needs to first determine the positive ideal solution and the negative ideal solution, and then compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain the score of each indicator parameter.

[0106] In this embodiment, the TOPSIS algorithm needs to first determine the positive ideal solution and the negative ideal solution, including:

[0107] Upper Bound Matrix and the lower bound matrix According to the standardized matrix Z = (z ij ) m×n Calculation results in formulas (8) and (9):

[0108]

[0109]

[0110] In the upper bound matrix ZU In, order is a positive ideal solution, where is the maximum value among the i-th index;

[0111] In the lower bound matrix Z L In, order z L ={ z 1 , z 2 ,... z m} T is a negative ideal solution, where z i is the minimum value of the i-th index, and T is the transposition operation.

[0112] It should be noted that the original TOPSIS method is aimed at problems with precise data. If there are complex data scenarios with precise values ​​and interval values, new measures need to be proposed. In addition, TOPSIS generally uses Euclidean distance to calculate the distance between alternative solutions (i.e., different parking spaces) and positive and negative ideal solutions. However, Euclidean distance may ignore the linear relationship between indicators, and when the evaluation results of alternative solutions are close, it is difficult to distinguish the advantages and disadvantages of different solutions. In order to improve the TOPSIS method, an upper bound matrix Z is established. U With the lower bound matrix Z L ; and then determine the positive ideal solution and negative ideal solution Z L . Introduce grey correlation to calculate the relationship between each parking space and the ideal solution and negative ideal solution Z L The TOPSIS method needs to first determine the positive ideal solution and the negative ideal solution, and then compare the indicator parameters of each solution with the positive and negative ideal solutions to obtain the score of each solution.

[0113] Optionally, a grey decision analysis algorithm is used to calculate the distance between each parking space and the positive ideal solution and the negative ideal solution, and the scores of each indicator parameter are ranked, including:

[0114] Upper bound matrix Z U The grey correlation matrix between the positive ideal solution is The elements in the matrix are:

[0115]

[0116] Lower bound matrix Z L The grey correlation matrix between the negative ideal solution is The elements in the matrix are:

[0117]

[0118] Among them, in formula (10) and formula (11), ρ represents the gray resolution coefficient, which is generally taken as 0.5;

[0119] The weighted grey correlation between each parking space and the positive ideal solution and the negative ideal solution calculated according to the grey correlation matrix is: The improved distance is:

[0120]

[0121] Among them, ω in formula (12) j is the attribute preference weight obtained in formula (4).

[0122] In this embodiment, the relative closeness T between the parking space and the positive ideal solution and the negative ideal solution is calculated. j As a quantitative scoring result, the relative closeness calculation formula is:

[0123]

[0124] When the relative closeness approaches 0, the closer the parking space is to the negative ideal solution, the lower the score; when the relative closeness approaches 1, the closer the parking space is to the positive ideal solution, the higher the score. The ranking of each parking space is obtained and a suitable parking space is recommended to the user.

[0125] See also Figure 2 The present invention also provides a parking space recommendation system based on the superiority and inferiority solution distance method, which is applied to the above-mentioned parking space recommendation method based on the superiority and inferiority solution distance method, comprising:

[0126] An index system construction module is used to construct an evaluation index system for driving travel scenarios, wherein the evaluation index system includes five parking space attributes, namely, the time to arrive at the parking space, the hourly parking price, the convenience of parking, the walking distance from the parking space to the final destination, and the safety of the parking lot;

[0127] An indicator parameter determination module, used to determine the attribute preference weights of parking spaces using the G1 algorithm according to the evaluation indicator system, and to sort the attribute preference weights to obtain indicator parameters of the evaluation indicator system;

[0128] An evaluation matrix forming module, used to collect and organize various attribute information of multiple parking spaces, and form a parking space multi-attribute evaluation matrix based on the various attribute information;

[0129] A positive and negative ideal solution determination module, used to establish an upper bound matrix and a lower bound matrix based on the parking space multi-attribute evaluation matrix, and respectively determine the positive ideal solution and the negative ideal solution corresponding to each of the indicator parameters, wherein the positive ideal solution and the negative ideal solution are determined by using a TOPSIS algorithm;

[0130] The parking space recommendation module is used to compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each indicator parameter, so as to recommend a suitable parking space for the user.

[0131] Specifically, two drivers are invited in advance to set the time period U for arriving at the parking space according to their preferences. 1 Hourly parking fee U 2 , Parking convenience 3 , walking distance from the parking space to the final destination U 4 、Parking lot safety 5 These five attributes are ranked and their importance ratios are determined i User 1's ranking is U 1 >U 2 >U 3 >U 4 =U 5 , and r 2 =1.8; r 3 =1.2; r 4 =1.4; r 5 =1.0. User 2's ranking is U 1 >U 2 >U 3 =U 4 >U 5 , and r 2 =1.6; r 3 =1.4; r 4 =1; r 5 =1.2.

[0132] Substituting the above information into formula (2) and formula (3), we can obtain the attribute weights of different users, as shown in Table 2:

[0133] Table 2 Weights of attributes for different users

[0134]

[0135] As can be seen from Table 2, user 1 has a certain time period U for reaching the parking space. 1 The weight given is the largest, followed by the hourly charge price U 2 and parking convenience 3 , indicating that user 1 is a time-sensitive driver, but he also cares about the parking fee and parking convenience. The parking space recommended for him must have advantages in terms of arrival time, price and parking convenience. User 2 is most concerned about the hourly parking fee U 2 , and secondly, the convenience of parking. 3, indicating that User 2 is a price-sensitive driver who also cares about the convenience of parking. The parking spaces recommended for him should be both low-priced and convenient.

[0136] Specifically, there are 8 parking spaces near the destination, and their information is shown in Table 3:

[0137] Table 3 Parking space information table

[0138]

[0139] The data in Table 3 can form a parking space multi-attribute evaluation matrix A = (a ij ) 5×8 , using formula (5) to formula (9), the matrix can be normalized and the upper bound matrix Z can be obtained U , and thus calculate the positive ideal solution:

[0140]

[0141] Get the lower bound matrix Z L , thus calculating the negative ideal solution

[0142] z L ={ z 1 , z 2 ,... z m} T ={0.1482,0,0.2387,0,0} T

[0143] Using formula (10) to calculate the upper bound matrix Z U The grey correlation matrix ξ between the positive ideal solution U ; Using formula (11), calculate the lower bound matrix Z L The grey correlation matrix ξ between the negative ideal solution L On this basis, the weighted grey correlation between each parking space and the positive and negative ideal solutions is calculated using formulas (12) and (13): and relative closeness T j , the data are shown in Table 4 and Table 5.

[0144] Table 4 Evaluation results of user 1

[0145]

[0146] As shown in Table 4, for user 1, the order of the eight parking spaces is 2>1>4>6>8>7>3>5. Parking space 2 has the highest score of 0.5417; parking space 1 has the second highest comprehensive score of 0.5083; parking space 3 has the lowest comprehensive score of 0.3835. Among the eight parking spaces, the time period for parking space 2 to arrive at the parking space is [25,38], which has its own advantages and disadvantages compared with the time period of parking space 3 [24,40], and is obviously better than the time period of parking space 1 [30,35]. However, user 1 is also concerned about the parking fee and parking convenience. In terms of parking convenience, parking space 2 is obviously better than parking space 3. Therefore, after comprehensive evaluation, parking space 2 is recommended for user 1.

[0147] Table 5 Evaluation results of user 2

[0148]

[0149] As shown in Table 5, for user 2, the order of the eight parking spaces is 1>2>6>4>7>8>3>5. Parking space 1 has the highest score of 0.5418; Parking space 2 has the second highest score of 0.5249; Parking space 5 has the lowest score of 0.3581; Parking space 1 has a significantly lower hourly parking fee than other parking spaces, and its parking convenience is also among the best; Parking space 2 does not have an advantage in the hourly parking fee indicator; Parking space 5 has a high fee and is difficult to park. Therefore, after comprehensive evaluation, parking space 1 is recommended for him.

[0150] In summary, this method can make full use of the objective information contained in the attribute data of parking spaces according to the user's preferences to comprehensively evaluate parking spaces. The comprehensive evaluation results of different parking spaces are quite different, which is convenient for comparing different parking spaces. Finally, this method can recommend suitable parking spaces to users.

[0151] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limiting, and thus other examples of the exemplary embodiments may have different values.

[0152] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0153] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A parking space recommendation method based on the superior and inferior solution distance method, characterized in that: The following steps are involved: Constructing an evaluation index system for driving travel scenarios, wherein the evaluation index system includes five parking space attributes: time to arrive at the parking space, hourly parking price, parking convenience, walking distance from the parking space to the final destination, and parking lot safety; Determine the attribute preference weights of parking spaces using the G1 algorithm according to the evaluation index system, and sort the attribute preference weights to obtain the index parameters of the evaluation index system; Collect and organize various attribute information of multiple parking spaces, and form a parking space multi-attribute evaluation matrix based on the various attribute information; An upper bound matrix and a lower bound matrix are established based on the parking space multi-attribute evaluation matrix, and a positive ideal solution and a negative ideal solution corresponding to each of the indicator parameters are determined respectively, wherein the positive ideal solution and the negative ideal solution are determined by using a TOPSIS algorithm; Each index parameter is compared with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each index parameter, so as to recommend a suitable parking space for the user.

2. The parking space recommendation method based on the superior and inferior solution distance method according to claim 1 is characterized in that: The attribute preference weights of parking spaces are determined using the G1 algorithm according to the evaluation index system, including: The differences in the importance that different users attach to parking space attributes are used as attribute preference weights, and the attribute preference weights are determined using the G1 algorithm; The evaluation index system is sorted, and the index parameters after sorting are recorded as I1≥I2≥,...≥I j-1 ≥I j , where I i ≥I j Indicates that the importance of indicator i is greater than or equal to the importance of indicator j; Defining Indicators I i-1 with I i The importance ratio is r i , then: Among them, ω i represents the weight of the i-th indicator; Summing k from 2 to n, we get: according to Calculate the relative weight of the last indicator as: Finally, we get ω through formula (4): i-1 The weights of the indicators are: oh i-1 =r i oh i-1 (4) Among them, i=2,3,...n-1,n.

3. The parking space recommendation method based on the superior and inferior solution distance method according to claim 1 is characterized in that: Collect and organize various attribute information of multiple parking spaces, and form a parking space multi-attribute evaluation matrix based on the various attribute information, including: Collect and organize m types of attribute information of n parking spaces to form a parking space multi-attribute evaluation matrix A = (a ij ) m×n ; The data in the parking space multi-attribute evaluation matrix is ​​normalized to obtain a standardized matrix Z=(z ij ) m×n ; When data a ij When is the exact value, When data a ij When is an interval value, After data processing, we get Among them, z ij represents the normalized data. represents the lower bound data after normalization, represents the upper bound data after normalization, Represents parking space multi-attribute information a ij The lower limit of Represents parking space multi-attribute information a ij The upper limit of 4. The parking space recommendation method based on the superior and inferior solution distance method according to claim 3 is characterized in that: The data in the parking space multi-attribute evaluation matrix is ​​normalized, including: Select the time to arrive at the parking space U1, the hourly parking price U2, the convenience of parking U3, the walking distance from the parking space to the final destination U4 and the safety of the parking lot U5; Interval values ​​are introduced to describe the uncertainty of the time to reach the parking space U1 and the convenience of parking U3. Precise values ​​are used to describe the hourly charging price of the parking space U2, the walking distance from the parking space to the final destination U4 and the safety of the parking lot U5. The convenience of parking U3 and the safety of the parking lot U5 are scored by experts on a scale of 1-9.

5. The parking space recommendation method based on the superior and inferior solution distance method according to claim 1 is characterized in that: The parking convenience of 1 indicates the most inconvenient parking, and the parking convenience of 9 indicates the most convenient parking. The driving travel scenarios include at least one of road driving, cruising to select a parking space, parking, and walking to the destination.

6. The parking space recommendation method based on the superior and inferior solution distance method according to claim 1 is characterized in that: An upper bound matrix and a lower bound matrix are established based on the parking space multi-attribute evaluation matrix, and the positive ideal solution and the negative ideal solution corresponding to each of the indicator parameters are determined respectively, including: Establish the upper bound matrix Z U With the lower bound matrix Z L , and determine the positive ideal solution and negative ideal solution Z L ; The grey decision analysis algorithm is used to calculate the difference between each parking space and the ideal solution. Negative ideal solution Z L distance; Among them, the TOPSIS algorithm needs to first determine the positive ideal solution and the negative ideal solution, and then compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain the score of each indicator parameter.

7. The parking space recommendation method based on the superior and inferior solution distance method according to claim 6 is characterized in that: The TOPSIS algorithm needs to first determine the positive ideal solution and the negative ideal solution, including: Upper Bound Matrix and the lower bound matrix According to the standardized matrix Z = (z ij ) m×n Calculation results in formulas (8) and (9): In the upper bound matrix Z U In, order is a positive ideal solution, where is the maximum value among the i-th index; In the lower bound matrix Z L In, order z L ={ z 1, z 2,... z m } T is a negative ideal solution, where z i is the minimum value of the i-th index, and T is the transposition operation.

8. The parking space recommendation method based on the superior and inferior solution distance method according to claim 7 is characterized in that: The grey decision analysis algorithm is used to calculate the distance between each parking space and the positive ideal solution and the negative ideal solution, and the scores of each indicator parameter are ranked, including: Upper bound matrix Z U The grey correlation matrix between the positive ideal solution is The elements in the matrix are: Lower bound matrix Z L The grey correlation matrix between the negative ideal solution is The elements in the matrix are: Wherein, in formula (10) and formula (11), ρ represents the gray resolution coefficient; The weighted grey correlation between each parking space and the positive ideal solution and the negative ideal solution calculated according to the grey correlation matrix is: The improved distance is: Among them, ω in formula (12) j is the attribute preference weight obtained in formula (4).

9. The parking space recommendation method based on the superior and inferior solution distance method according to claim 8, characterized in that: Also includes: The relative closeness T between the parking space and the positive ideal solution and the negative ideal solution j As a quantitative scoring result, the relative closeness calculation formula is: When the relative closeness approaches 0, the closer the parking space is to the negative ideal solution, the lower the score; when the relative closeness approaches 1, the closer the parking space is to the positive ideal solution, the higher the score. The ranking of each parking space is obtained and a suitable parking space is recommended to the user.

10. A parking space recommendation system based on the superior and inferior solution distance method, characterized in that: The parking space recommendation method based on the superior-inferior solution distance method as described in any one of claims 1 to 9 comprises: An index system construction module is used to construct an evaluation index system for driving travel scenarios, wherein the evaluation index system includes five parking space attributes, namely, the time to arrive at the parking space, the hourly parking price, the convenience of parking, the walking distance from the parking space to the final destination, and the safety of the parking lot; An indicator parameter determination module, used to determine the attribute preference weights of parking spaces using the G1 algorithm according to the evaluation indicator system, and to sort the attribute preference weights to obtain indicator parameters of the evaluation indicator system; An evaluation matrix forming module, used to collect and organize various attribute information of multiple parking spaces, and form a parking space multi-attribute evaluation matrix based on the various attribute information; A positive and negative ideal solution determination module, used to establish an upper bound matrix and a lower bound matrix based on the parking space multi-attribute evaluation matrix, and respectively determine the positive ideal solution and the negative ideal solution corresponding to each of the indicator parameters, wherein the positive ideal solution and the negative ideal solution are determined by using a TOPSIS algorithm; The parking space recommendation module is used to compare each indicator parameter with the positive ideal solution and the negative ideal solution to obtain a score corresponding to each indicator parameter, so as to recommend a suitable parking space for the user.

Images